Static Generation of Efficient OpenMP Offload Data Mappings
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916385892859904 |
|---|---|
| author | Marzen, Luke Dutta, Akash Jannesari, Ali |
| author_facet | Marzen, Luke Dutta, Akash Jannesari, Ali |
| contents | Increasing heterogeneity in HPC architectures and compiler advancements have led to OpenMP being frequently used to enable computations on heterogeneous devices. However, the efficient movement of data on heterogeneous computing platforms is crucial for achieving high utilization. Programmers must explicitly map data between the host and connected accelerator devices to achieve efficient data movement. Ensuring efficient data transfer requires programmers to reason about complex data flow. This can be a laborious and error-prone process since the programmer must keep a mental model of data validity and lifetime spanning multiple data environments. We present a static analysis tool, OMPDart (OpenMP Data Reduction Tool), for OpenMP programs that models data dependencies between host and device regions and applies source code transformations to achieve efficient data transfer. Our evaluations on nine HPC benchmarks demonstrate that OMPDart is capable of generating effective data mapping constructs that substantially reduce data transfer between host and device. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_13881 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Static Generation of Efficient OpenMP Offload Data Mappings Marzen, Luke Dutta, Akash Jannesari, Ali Distributed, Parallel, and Cluster Computing Increasing heterogeneity in HPC architectures and compiler advancements have led to OpenMP being frequently used to enable computations on heterogeneous devices. However, the efficient movement of data on heterogeneous computing platforms is crucial for achieving high utilization. Programmers must explicitly map data between the host and connected accelerator devices to achieve efficient data movement. Ensuring efficient data transfer requires programmers to reason about complex data flow. This can be a laborious and error-prone process since the programmer must keep a mental model of data validity and lifetime spanning multiple data environments. We present a static analysis tool, OMPDart (OpenMP Data Reduction Tool), for OpenMP programs that models data dependencies between host and device regions and applies source code transformations to achieve efficient data transfer. Our evaluations on nine HPC benchmarks demonstrate that OMPDart is capable of generating effective data mapping constructs that substantially reduce data transfer between host and device. |
| title | Static Generation of Efficient OpenMP Offload Data Mappings |
| topic | Distributed, Parallel, and Cluster Computing |
| url | https://arxiv.org/abs/2406.13881 |